Political Odds

I have more of an interest in all the kerfuffle leading up to the election than the actual election itself… which is either a sad reflection on political apathy, or of my love of markets and probabilities. I’m not sure which.

That interest was somewhat dampened, it must be said, by having had the misfortune of watching the BBC Panorama program a couple of weeks ago where Nate Silver (the renowned US statistician) trucked around dreary English towns in a pantomime-esque attempt to drag out a forecast of the UK election.

That aside, the game of trying to figure out what will happen in the next election is a fascinating one from the point of view of a market watcher. London related shares moved when Labour announced their proposals around res non-doms (a nice tax structure for wealthy ‘foreigners’, basically), which investors saw as a threat to the status quo at the top end of the London market. The journalists also have a field day, since it gives them a nice event they can attribute market movements to – ‘the FTSE dropped 20 points as fears of a Labour-led Government grew..’ and the like.

So how to think about what is likely to happen?

As always, the best thing to do is probably to follow the money. For those who aren’t familiar with Betfair, it’s a betting site whereby the bets you place aren’t taken up by the house (which is adversarial – when you win, they lose), but instead by other people. It’s a functioning market, where the opinions of participants are aggregated to create at price at which you can back the bet – take the bet on – and a price at which you can lay it – bet that the outcome will not happen.

It also has spreads AIM would be jealous of.

Political Odds

Interestingly, you can also see how odds have changed historically – how the market is perceiving outcomes over time. The following is a chart of the implied likelihood of a Labour minority Government (with decimal odds, you can calculate implied probabilities by simply doing 1/x, where x is decimal figure).

Lab Minority

Historically, these markets prove much better predictors of actual outcomes than polls leading up to the election, or the opinions of political commentators. This isn’t really very surprising – the reason the stock market works fairly well (most of the time) is that, if there’s a large golden carrot in pricing things better than other people, lots of smart people will devote time and resources to making sure pricing is being done in the most rational way.

Anyway, if you’re feeling particularly spiteful, there’s probably money to be made by gaming your friends and co-workers into betting with you on their preferred party. People always overstate the likelihood of their pet outcome winning!

As for me – I suspect the election will be a bit of a damp squib. I think people overstate the impact of elections on equity markets, and I also think the political gesturing before the election makes people far more paranoid about what will happen than they need to be. A lot of hot air gets blown in the race, and you’ll forgive me for taking the cheap shot when I say that, frankly, most of that hot air ends up billowing harmlessly into the air when the politicians actually get to work.

Famous last words.

2 Replies to “Political Odds”

  1. Aktieingenjören

    I’ve seen similar claims but I am not sure on if betting stories are a bit of the “nice anecdote” variety which makes people forget when the pollsters are actually right.

    But it is a bit ironic that I have actually made some money (50 £) on political betting. The Swedish betting market is rather conservative and focused on horses and football. But we got major growth from online betting where Swedish companies such as Betsson and Unibet have tried to establish other forms of betting as well. Six months before the 2010 elections the polls where very close between the left wing block and the right wing block, so several betting companies ran campaigns where they had 1,96 times the money if you could guess what side would win.

    Which of course was very silly as we got the same left-leaning bias in polls and a historical tendency for the right-leaning parties to do well in the last months before elections (statistics and in depth political debate seem to have a right-wing bias in Sweden…). But my guess is that the betting firms where fairly confident that a mixture of left-wingers and media coverage would offset the cost of such favourable odds.

    • Lewis

      It’s a bit of a market efficiency argument, I guess.

      I’m with you on polls and other market sources tending to be biased – and this is a compounded problem from an individual perspective, because most of us live in bubbles which tend to over-represent the point of views we hold. Left-wing Guardian reading chaps are more likely to see the polls which tend to favour the left-wing parties than avid UKIPpers, who’ll log onto Twitter and see plenty of messages from their friends talking about how well UKIP are doing in X or Y seat. So you get biased polled compounded by looking at things from a certain lens.

      I think betting markets are much better incorporators of that information than almost any individual is – in the same way that the market as a whole prices stocks much better than I do. On that same basis, though, I can make money both on the markets and betting (fingers crossed for continuance…) when you identify discrepancies – which do, as you point out, exist!

      The odds have moved massively since I posted this article – the odds of a Conservative minority govt. having gone from 15 to 20%, for instance. I’m not entirely sure whether that is an argument for the wisdom of using a market which is continually pricing changes and is hence your best bet… or a sign that no-one really has any clue, and the market is more akin to what penny stocks do when you leave them with a bunch of bored punters!

Leave a Reply

Your email address will not be published. Required fields are marked *